Detection Method of Large Coal Blocks on Conveyor Belt Based on CA-YOLOv8

To address the problem of inaccurate detection and poor generalization of large coal blocks caused by image blur due to coal dust interference, uneven lighting, and high-speed movement of conveyor belts, a method for detecting large coal blocks based on CA-YOLOv8 is proposed. Firstly, image restoration is used to preprocess the blurred images of the conveyor belt, which restores the image to be clear and provides high-quality image input for subsequent large coal block detection. Then, the YOLOv8 algorithm is improved according to the actual monitoring scenario, including adaptive image scaling based on real size at the Input, embedding the coordinate attention in the network structure, and removing unnecessary detection heads, which achieves accurate, stable and efficient detection of large coal blocks. The experimental results prove that the method in this paper can not only accurately detect large coal blocks in blurred images, but also exhibits good generalization ability under different environmental conditions.